“Critical Notice of G.A. Cohen’s Self-Ownership, Freedom, and Equality”, Canadian Journal of
Philosophy 28 (1998): 609-626.
Peter Vallentyne
SELF-OWNERSHIP FOR EGALITARIANS
G.A. Cohen’s book brings together and elaborates on articles that he has written on self-
ownership, on Marx’s theory of exploitation, and on the future of socialism. Although seven of
the eleven chapters have been previously published (1977-1992), this is not merely a collection
of articles. There is a superb introduction that gives an overview of how the chapters fit together
and of their historical relation to each other. Most chapters have a new introduction and often a
postscript or addendum that connect them with other chapters. And the four new chapters (on
justice and market transactions, exploitation in Marx, the concept of self-ownership, and the
plausibility of the thesis of self-ownership) are important contributions that round out and bring
closure to many of the central issues. As always with Cohen, the writing is crystal clear, and full
of compelling examples, deep insights, and powerful arguments.
Cohen has long been recognized as one of the most important exponents of analytic
Marxism. His innovative, rigorous, and exciting interpretations of Marx’s theories of history and
of exploitation have had a major impact on Marxist scholarship. Starting in the mid-1970s he
has increasingly turned his attention to normative political philosophy. As Cohen describes it, he
was awakened from his “dogmatic socialist slumbers” by Nozick’s famous Wilt Chamberlain
example in which people starting from a position of equality (or other favored patterned
distribution) freely choose to pay to watch Wilt Chamberlain play, and the net result is inequality
(or other unfavored pattern). During the subsequent twenty years, political philosophy has
2
benefited from his thinking about the nature and plausibility of the thesis of self-ownership, and
about the scope and demands of equality.
In what follows I will focus solely on the material dealing with self-ownership, but first I
shall mention some of the interesting material on Marxism and socialism that I will be ignoring.
First, at various points Cohen discusses how something like a principle of self-ownership is
latent in the standard Marxist condemnation of capitalist exploitation (e.g., the capitalist steals
labor time from the laborer). It is precisely because of this that Cohen has taken self-ownership
so seriously. Second, Cohen discusses the labor theories of value of Marx and Locke and their
connection to Marx’s theory of exploitation. Third, Cohen discusses how contemporary social,
economic, and political circumstances make political change much more complex than is used to
be for Marxists and socialists. In the early days of Marxism, workers (in the sense of
“employees”) as a group made up a majority of the population, and were roughly coextensive
with the group of needy people and with the group of exploited people. But things are more
complex today, because of the existence of many more stockholding workers, well-paid workers
(e.g., managers), and needy non-workers (e.g. poor unpaid homemakers and semi-permanent
unemployed people). Finally, Cohen criticizes Marx’s “technological fix” for the problem of
justice (roughly: that technological advances will inevitably produce abundance and thereby
eliminate the circumstances of justice), and defends a moderate “social fix” according to which
under conditions of moderate scarcity people over time can become disposed to comply with the
demands of justice.
I shall focus solely on the issue of self-ownership—both because it is the dominant topic
of the book, and because it is the topic that I am most qualified to comment on. I shall start by
rehearsing and endorsing Cohen’s analysis of the thesis of self-ownership. Then I shall identify
3
and motivate an egalitarian view that endorses self-ownership, and which is not refuted by
Cohen’s criticisms.
1. The Concept of Full Self-Ownership
Cohen gives three more or less equivalent characterizations of universal full self-ownership:
[E]ach person enjoys over herself and her powers full and exclusive rights of control and use, and
therefore owes no service or product to anyone else that she has not contracted to supply. (p. 12)
[E]ach person possesses over himself, as a matter of moral right, all those rights that a
slaveholder has over a complete chattel slave as a matter of legal right, and he is entitled, morally
speaking, to dispose over himself in the way that such a slaveholder is entitled, legally speaking,
to dispose over his slave. (p. 68)
[E]ach person has an extensive set of moral rights (which the law of his land may or may not
recognize) over the use and fruits of his body and capacities, comparable in content to the right
enjoyed by one who has unrestricted private ownership of a piece of physical property. (p. 117)
The core idea is that agents own themselves in just the same way that they can have
maximal private ownership in a thing. This maximal private ownership is typically taken to
include the right to fully manage (to use, and to allow or prohibit others from using), the right to
the full income; the right to transfer fully any of these rights through market exchange, inter
vivos gift, or bequest; and the right to recover damages if someone violates any of these rights.
Redistributive taxation (e.g., of income or wealth) is incompatible with these rights of maximal
private ownership.
4
Throughout, we will understand the claim of full self-ownership to be implicitly
conditional upon the agent’s not having committed any previous wrongs (which may lead to
partial or full loss of self-ownership) and upon the agent’s not having contractually given up
some of his/her self-ownership. No one defends universal unconditional full self-ownership.
We should also note that claims of self-ownership can be understood in an ethical sense
and in a political sense. Ethical self-ownership concerns the moral rights correlative to the moral
duties of others. Political self-ownership concerns the legitimately enforceable moral rights
correlative to the legitimately enforceable moral duties of others. The political conception is
stronger than the ethical conception in that it asserts that one’s rights of self-ownership are
legitimately enforceable, and it is weaker in that it does not rule out having various non-
enforceable non-contractual duties to others. My discussion will focus on the political reading,
and for brevity references to rights/obligations should be understood throughout as references to
the legitimately enforceable obligations. Likewise references to morality should be understood as
references to the legitimately enforceable part of morality (justice).
Finally, to keep things simple, we’ll assume that all beings with moral standing are fully
psychologically autonomous (fully capable of self-governance), and that they pop into existence
in this state (as opposed to developing it gradually). The existence of non-human animals,
children, non-autonomous adults, and the like introduces a number of extremely important
complications that cannot be addressed here.
2. Some Bad Rationales for Full Self-Ownership
In Ch. 9 (new) Cohen argues that several considerations which are often taken to support the
thesis of full self-ownership do not in fact support it. As he acknowledges, these are only
5
criticisms of supporting arguments, and not criticisms of the thesis itself. Still, the supporting
arguments are important, and their failure casts some doubt on the thesis itself.1
One argument in favor of full self-ownership (crudely put) is this: (1) If we are not full
self-owners, then we are at least partial slaves. (2) We are not partial or full slaves. So, (3) we
are full self-owners. Cohen insightfully points out that the first premiss is false. For the missing
rights of full self-ownership may not be held by anyone in the appropriate manner to constitute
partial ownership of another. For example, one may have a duty to help another (and thus have
incomplete self-ownership), without the other having a right to that help, the right to waive the
help, or other ownership rights over the other. Of course the issue is complex, and depends on
whether all duties are based on rights held by someone, and on what sorts of rights constitute
partial ownership, but Cohen is clearly right that lack of full self-ownership does not entail
partial slavery.
As Cohen makes clear, most defenders of full (or even partial) self-ownership will reply
that even if there is no partial slavery, the relevant issue is whether one is appropriately subject to
one’s own will. So a second line of defense of full self-ownership is to appeal to the importance
of autonomy (understood roughly as the effective ability to control one’s life as determined by
the morally permissible range of choices). But, as Cohen argues, self-ownership does not always
support autonomy. First, for those who are poorly endowed, universal full self-ownership may
leave them less effective autonomy than some scheme of universal partial self-ownership (e.g.,
where everyone has a duty to provide the needy with the means of subsistence). Second, even if
everyone is equally endowed, universal full self-ownership may provide less autonomy for each
than an alternative scheme (e.g., public goods may be under-supplied and thereby fail to increase
everyone’s autonomy). Finally, and most importantly, self-ownership is a purely formal notion,
6
and does not ensure that one has any effective autonomy involving physical actions. If the rest of
the world is owned by others, no physical action is permitted without someone else’s approval
(since anything one does would involve using someone else’s property). In such a situation one
has no effective autonomy involving physical actions.
So, self-ownership does not ensure (effective) autonomy. It is concerned instead with
ensuring that there are certain formal constraints on what others may do to one without one’s
consent, rather than with ensuring that one has the means to do things.
A third line of defense of full self-ownership is the following: (1) Forced transfer of body
parts (e.g., removing someone’s eye to transfer to a blind person) is unjust under current
technological conditions. (2) Full self-ownership holds that it is unjust. (3) No theory that
rejects full self-ownership can adequately account for the injustice. So, (4) full self-ownership is
true.
Cohen questions the third premiss. He asks us to consider a case where (starting at some
point) everyone is born without eyes, and there is a well-established, state-funded practice of
perinatal implantation of perfect state-owned artificial eyes. Sometimes the eyes of adults cease
to function (e.g. because of accidents), and when both eyes cease to function they are supplied
with an eye from another adult with two good eyes, who is selected by lottery. (Adult implants
work only if the eye has been in constant use from birth by some adult.) The mandatory transfer
of the eye is not a violation of the “donor’s” self-ownership, because the eyes are owned by the
state and only loaned to the users, with it being understood that they may have to give up an eye
if their name is drawn in the lottery.
If the mandatory transfer in this hypothetical case is as morally problematic as a real life
mandatory transfer, Cohen suggests, it is because of the gross interference in the “donor’s” life.
7
And since this gross interference is also present in the real life cases, there is no need to posit full
self-ownership to explain why such transfers are unjust. The third premiss is false.
As Cohen acknowledges, defenders of full self-ownership will be likely be unpersuaded.
For they will likely deny the moral equivalence of the two cases. In both cases there is significant
disruption in the “donor’s” life. But in the hypothetical case, defenders of self-ownership will
claim, this is unfortunate but not unjust (it was part of the terms of the loan agreement), whereas
in the typical real life case it is both unfortunate and unjust (it involves coercively taking
something that rightly belongs to a person). Still, Cohen’s discussion insightfully highlights how
one can condemn forced transfer of body parts without invoking self-ownership.2
So far we have considered some criticisms of some arguments in favor of full self-
ownership. Of course, the unsoundness of several arguments for full self-ownership does not
refute the thesis. Let us now turn to some criticisms of the thesis itself.
3. Cohen Against Full Self-Ownership
Full self-ownership says nothing about the ownership of the non-agent part of the world: natural
resources (unproduced resources) and artifacts (produced resources). Cohen very insightfully
explores the implications of combining full self-ownership with various views about the
ownership of the rest of the world. He concludes that the thesis of full self-ownership must be
rejected primarily because it is incompatible with adequate protection of autonomy combined
with adequate promotion of equality of condition.3
Libertarianism is the view that agents are full self-owners. The most common form is
right-libertarianism, which combines full self-ownership with the view that the rest of the world
is initially unowned, or held in common, and that portions come to be privately owned as agents
8
mix their labor with, or first occupy, or perform some other type of activity on, portions of the
unowned world, perhaps conditional on some sort of minimal compensation conditions (e.g., that
no one be made worse off). Cohen rightly criticizes this view for failing to adequately promote
equality of condition and for failing to adequately protect effective autonomy (since a self-owner
who owns nothing else has little effective autonomy).
Left-libertarianism endorses full self-ownership, but holds that natural resources
(understood as including the natural resource component of artifacts) are owned in some
egalitarian sense. It might seem that egalitarians can endorse full self-ownership by endorsing
left-libertarianism. Cohen argues, however, that each of the main versions left-libertarianism is
either insufficiently egalitarian or gives insufficient protection to autonomy.
According to one version of left-libertarianism, natural resources are jointly owned in the
sense that all decisions about use, etc. are to be made through a collective decision-making
process. One form of this view holds that all decisions are to be made by unanimous agreement.
Although this view would presumably guarantee rough equality of condition (since everyone
must agree to any arrangement), Cohen rightly rejects it on the grounds that no one would have
any effective autonomy. For although it holds that each person is a full self-owner, no one has
the right to do anything (e.g., stand in a given spot, eat an apple, or even breathe) without
authorization from every other person. This is because every action requires the use of some
natural resources (e.g., to occupy a spatial location), and so no one is permitted to do anything
without approval from others.
Although Cohen (rightly) rejects the joint ownership of natural resources because it fails
to ensure a minimal amount of autonomy for each person, he notes that right-libertarians cannot
reject it on these grounds. For right-libertarianism does not guarantee that people have any
9
effective autonomy. It just insists on the formal autonomy guaranteed by self-ownership (which
the above version of left-libertarianism also insists on). It sees no problem, for example, with a
situation in which one person (or group of people) through luck or hard work comes to own all
non-agent resources, so that all others require her permission to do anything.
Of course, collective decision-making need not require unanimity. It might only require a
majority. But this version of joint ownership of natural resources would neither guarantee
equality of condition (since a majority could decide to exclude others from benefits), nor leave
agents any effective autonomy (since they would still require authorization from others for every
action). So this is not a plausible way of combining full self-ownership with social ownership of
natural resources either.
A second version of left-libertarianism holds that natural resources are equally owned by
all in the sense that everyone is entitled to an equally valuable share of the natural resources.
This avoids the above problem of collective authority over the natural resources, and thus over
individual choices, because it divides up natural resources into equally valuable private shares.
The most natural way of evaluating different shares of natural resources is in terms of their
competitive value determined by supply and demand under relevant market conditions (e.g., their
price at an auction). Each person would owe a tax equal to the value of any natural resources that
he/she has appropriated, and the social fund so generated would then be divided equally among
the members of society. If there is only one generation of agents, then agents could be charged
the purchase price of the natural resources that they have appropriated. If there are many
generations, then a more natural approach is to charge agents periodic rent for the natural
resource that they have appropriated.4
10
Cohen rightly rejects this second approach primarily because it does nothing to
compensate for inequalities in brute luck (e.g., in initial personal endowments or later events).
Because each person gets an equal share of the social fund, and full-self-ownership prohibits
coercive redistributive taxation, nothing is done to help alleviate the burdens of bad brute luck.
On the basis of his examination of these two approaches Cohen concludes that no
combination of full self-ownership and egalitarian ownership of natural resources adequately
ensures both the protection of autonomy and the promotion of equality. Although I agree with
this conclusion, further argument is needed before it is reasonable to draw this conclusion. For
there are other, more plausible, conceptions of egalitarian ownership of natural resources with
which full self-ownership can be combined. Below, I shall identify some such conceptions. First,
however, I will discuss why egalitarians should endorse some form of self-ownership.
4. The Need for Some Form of Self-Ownership
Some form of self-ownership is, I claim, a plausible constraint on the demands of equality.
Unconstrained teleological egalitarianism holds that equality should be maximally, or at least
adequately, promoted. Under some circumstances, it therefore judges it just (and perhaps
required by justice) to torture, kill, or involuntarily enslave innocent people. It will do so
whenever such treatment is the most effective means to the end of equality (e.g., if it will make
certain resources available for the purposes of equalization). But it is unjust to torture, kill, or
involuntarily enslave innocent people. The ends do not always justify the means. Hence
unconstrained teleological egalitarianism is mistaken. Obviously, the issue is complex and
controversial, and I am here simply outlining—rather than developing—an argument.
11
Egalitarianism must recognize some constraints on the admissible means of promoting
equality. The most plausible grounds for these constraints are, I think, generated by some form of
self-ownership and some account of ownership of natural resources. In particular, some form of
self-ownership is necessary to adequately recognize that there are some things that others may
not do to an agent without his/her consent (e.g., various forms of physical contact), and that those
very things are permissible if the agent gives his/her consent and the owners of other involved
resources give their consent. Again, this is highly controversial. A constraint against torture need
not be accompanied by a right to waive the constraint, as self-ownership holds. But only
waivable constraints, I would argue, adequately recognize the formal rights of control that agents
have over the use of their persons.
The assumption that some form of self-ownership imposes constraints on equality
promotion need not be the assumption that full self-ownership imposes such constraints. Full
self-ownership gives agents various control rights over the use of their persons. But it also gives
them rights to transfer those rights to others, and tax immunities for the possession and exercise
of these rights. One can endorse a partial form of self-ownership (e.g., control rights) without
endorsing full self-ownership (with full tax immunities).5 Indeed, the unrestricted right to transfer
tax immune wealth by gift must, I believe, be rejected by egalitarians—at least for cases of
wealth being transferred to a person of a younger generation. For without some such restriction,
wealth dynasties will be generated, and no plausible form of egalitarianism can accept those.6 For
simplicity, however, I shall ignore this concern below, and focus on how full self-ownership can
be combined with a plausible conception of the ownership of natural resources. If full self-
ownership is so combinable, then so are weaker forms.
12
5. The Ownership of Natural Resources Revisited
As Cohen recognizes, the significance and plausibility of self-ownership can be assessed only in
the context of plausible assumptions about the ownership of natural resources. For full self-
ownership has no substantive implications on its own. It all depends on its interplay with the
ownership of natural resources. The rights of full self-ownership are conditional upon not having
voluntarily given them up and upon not having committed previous wrongs. And this means that
whatever rights full self-ownership “initially” bestows, they may—depending on the terms of
ownership of natural resources—be immediately lost because of the use of natural resources that
belong in part to others, or because of the terms governing the legitimate appropriation of
unappropriated resources. If, for example, all natural resources are owned by others (e.g.,
privately by some individual or jointly by society), then agents may own none of their products
made from natural resources (they may involve theft and trespass), and they may owe others
services or products as compensation for violation of their property rights. So, self-ownership can
only be fully assessed as part of a package that includes a plausible view about the ownership of
natural resources.
As Cohen’s discussion of joint ownership of natural resources shows, a plausible
conception of the ownership of natural resources should allow unappropriated resources to be
used by agents without the permission of others and without any loss of the rights of self-
ownership. More specifically, it should be common-use-based in the sense that (roughly) agents
are permitted to use an unappropriated natural resource as long as they violate no one’s self-
ownership, no one else is currently using it, and perhaps subject to certain other conditions.7
A plausible conception of the ownership of natural resources should also be unilateralist
in the sense of allowing appropriation of unappropriated natural resources without the consent of
13
others—as long as certain conditions (to be discussed below) are met. For it is most implausible
to hold that the consent of others is required when communication with others is impossible or
extremely difficult or expensive (as it almost always is). And even when communication is
relatively easy and costless, it’s unclear why one needs the consent of others as long as one
makes an appropriate payment for the natural resources appropriated.
We have already considered three common-use-based, unilateralist conceptions of natural
resource ownership (in conjunction with full self-ownership). Radical right libertarians—such as
Rothbard (1978, 1982) and Kirzner (1978)—hold that that there are no compensation conditions
on the appropriation of unappropriated resources. Agents are free to take ownership of whatever
unappropriated natural resources they find (or mix their labor with). Obviously, this is a non-
starter from an egalitarian viewpoint. Lockean right libertarians (such as Nozick 1974) hold that
the only condition is a Lockean proviso, which requires roughly that no individual be made
worse off (in some appropriate sense) by the appropriation. It seems quite plausible that
satisfaction of some form of a Lockean proviso is a necessary condition for just unilateral
appropriation. But Lockean libertarians are mistaken in holding that it is sufficient for just
appropriation. For private property rights over natural resources typically bring the owners
benefits (even after making payments to ensure that no one is made worse off). Consequently,
people are willing to pay for these rights, and the rights have—relative to the initial condition of
self-ownership, common use, and the obligation to compensate anyone made worse off by
appropriation—a competitive value (market clearing price).8 Given the valuable nature of rights
over natural resources, there is no reason why an appropriator should pay less than this
competitive value. This view, in conjunction with full self-ownership was developed and
defended by Henry George (1879, 1892) and others, and is known as Georgist libertarianism.9
14
Given the existence of multiple generations, the most plausible version of this approach is to
require that rights over natural resources be rented at the competitive rent value. For if the natural
resources were purchased at their competitive value, and these funds were allocated among the
existing members of society, members of future generations would be deprived of the value of
the natural resources.
Equal share left-libertarianism (e.g., Steiner’s theory), considered above, is a version of
Georgist libertarianism. It holds that the competitive rents should be divided equally among all
members of society. Cohen rightly criticized this view for not being adequately egalitarian.
Because it divides the social fund equally, no special compensation is provided to those who
suffer unchosen disadvantage. The commitment to equal division, however, is not a necessary
commitment of Georgist libertarianism (which only requires full self-ownership and the payment
of competitive rent for appropriation). A more plausible view, I think, is an equality-promoting
Georgist conception, according to which the social fund from rents is spent to promote equality
of unchosen advantage. It focuses spending on the disadvantaged rather than dividing the fund
equally. It is thus significantly more egalitarian in terms of equality of effective opportunity for a
good life.10
This conception is compatible with full self-ownership. A main attraction of this
combination is, I think, that it restricts the demands of equality to the spending of the social fund.
The size of the social fund is determined by the competitive value of natural resource
appropriated—and not by the amount needed to eliminate unchosen disadvantage. This view
treats natural resources as resources to promote equality, but it gives clear recognition to the
normative separateness of persons, and places clear—and arguably plausible—limits on our
duties to others.
15
Cohen and most egalitarians, however, would reject equality-promoting Georgism on the
grounds that the taxes (rents) that it requires agents to pay are based solely on the competitive
value of the natural resources that they own. As a result, agents with more advantageous
unchosen personal endowments (e.g. productive talents) pay the same taxes as those with less
advantageous unchosen personal endowments who own equally valuable natural resources. Most
egalitarians will agree with the spending policy of equality-promoting Georgism, but they will
reject the Georgist taxation policy that limits taxes to the competitive value of natural resources
appropriated. Most egalitarians will want those with greater unchosen advantage to pay higher
taxes. Doing this will further promote equality—both by the leveling down effect of imposing the
taxes and by the leveling up effect from spending them.
Most such egalitarians may grant that the payment of competitive rent is a necessary
condition for just appropriation, but they will deny that it is a sufficient condition. A natural way
of modifying the Georgist position to take into account the above consideration is to hold that, in
addition to paying the competitive rent, appropriators must pay a tax equal to up to 100% of the
net benefits (net of the competitive rent) that they reap from the appropriation. Of course, in
practice it is not viable to tax agents 100% of the benefits they reap from the appropriation of
natural resources. The required information about benefits is impossible to obtain and even
enough information for rough approximations would be very costly to obtain. Furthermore, full
benefit taxation leaves no incentive to make productive use of natural resources (since it leaves
no net benefit to the agent). For these reasons, the full benefit taxation approach should be
understood as setting a maximum tax that can be charged. The actual tax charged will be
whatever maximizes net tax revenues (after deducting administrative expenses).
16
Consider then the equality-promoting, full benefit taxation conception of natural resource
ownership. This is like the equality-promoting Georgist view considered above except that, in
addition to paying competitive rent, appropriators must pay taxes (up to 100%) on the benefits
they reap from appropriation. This approach has the effect of treating all benefits of applying
personal talents to appropriated natural resources as a social asset. It is, however, compatible
with full self-ownership. First, as we have seen, any assumption about the ownership of natural
resources is compatible with full self-ownership. Second and more importantly, this view of the
ownership of natural resources is compatible with a relatively secure self-ownership. For, like the
Georgist conception, it imposes the obligation to pay the benefit taxation only on those who
appropriate natural resources. Agents are free to use unappropriated natural resources under the
terms of common use without acquiring any obligation to pay rent or benefit taxes. It is thus
possible for agents to avoid having to pay the tax.11
Equality-promoting, full benefit taxation libertarianism should be attractive to many
egalitarians. It is not, however, a form of pure egalitarianism. For its endorsement of full self-
ownership does some real work in limiting the admissible ways in which equality may be
promoted. Agents may not be killed, tortured, or assaulted without their consent. Nor may they
be coerced into providing involuntary services for others (e.g., mandatory labor for the state). Nor
do agents owe any taxes merely because they exist or because they use natural resources. Nor are
taxes imposed that effectively require agents to work in their most productive capacity (e.g., the
“tax slavery” that results if each person owes a tax equal to the value of his/her maximally
valuable annual product). If, however, agents appropriate natural resources, then they must pay
the competitive rent plus taxes equal to up to the full value of the net benefit from appropriation
(net of competitive rent).
17
Of course, more demanding (less constrained) conceptions of the ownership of natural
resources are possible. Purist egalitarians could hold that anyone who uses natural resources
thereby incurs the obligation to do whatever is necessary to maximally promote equality.
Although such an approach is formally compatible with (initial) full self-ownership, it gives no
real role to self-ownership, since agents must use some natural resources (e.g., to stand on or to
breathe), and hence immediately lose their self-ownership. More weakly, one could hold that the
obligation to promote equality maximally is incurred by anyone who appropriates (as opposed to
uses) natural resources. This leaves some real role for self-ownership, since in principle agents
could decide not to appropriate. But it has the result that appropriators may be subject, under
certain conditions, to involuntary service (e.g., when their skills are needed by society), forced
transfer of body parts (e.g., from an advantaged person to a disadvantaged person), or even
torture (e.g., when it provides important information that reduces the suffering of the
disadvantaged). Because these implications seem implausible, nothing more demanding than
equality-promoting full benefit taxation (or something like it) seems promising for egalitarians.12
More generally, something in the general area between equality-promoting Georgism and
equality-promoting full benefit taxation seems promising for egalitarians. Each is compatible
with full self-ownership and is more plausible than joint-ownership and than equal-share
Georgism. They each avoid the problem of requiring the consent of others to use natural
resources by holding that common use is permitted and involves no waiving of any rights of self-
ownership. They also hold that appropriation of natural resources without the consent of others is
legitimate as long as appropriators pay the relevant rent and taxes. Georgism requires agents to
pay competitive rent, and thus typically allows agents to benefit from appropriation, with agents
18
with greater productive capacities typically reaping greater benefits. Full benefit taxation, on the
other hand, taxes away up to the full benefit of appropriation.
If equality of unchosen advantage is the only concern of justice, then both equality-
promoting Georgism and equality-promoting full benefit taxation are flawed (because
inadequately egalitarian). Most of Cohen’s writing (e.g., Cohen 1989) make it seems that Cohen
is committed to a pure form of egalitarianism, and if he is, then he will reject both these views.
But in places (e.g., his rejection of joint ownership of natural resources, and his endorsement of
the idea of an egalitarian bill of rights) he seems to favor some sort of constrained version of
egalitarianism. For those favoring this latter approach, both equality-promoting Georgism and
equality promoting full benefit taxation are, in conjunction with self-ownership, promising
views. The key difference between the two concerns whether agents are entitled to the net
benefits of their labor (net of competitive rent). Equality-promoting Georgism says they are, and
equality-promoting full benefit taxation say that they aren’t. I shall not attempt here resolve this
issue. My main claim is simply that something in this general area is a promising conception of
the ownership of natural resources for non-purist egalitarians.
This is not to say, however, that egalitarians should endorse full self-ownership. So far I
have simply assumed full self-ownership and explored some conceptions of the ownership of
natural resources. As suggested in the previous section, I believe that full self-ownership is too
strong in that it gives agents the right to transfer tax immune wealth by gift (and bequest) to
members of later generations. Such a right can give rise to wealth dynasties, and that is
unacceptable from an egalitarian perspective. Egalitarians need, I suggest, to endorse a form of
self-ownership, but it need not—and should not—include that right to make untaxed gifts. But
that’s a topic for another occasion.13
19
6. Conclusion
Cohen’s discussion of self-ownership is extremely insightful and important. I have not here
addressed all the criticisms that he raises against full self-ownership. I have focused on his claim
that full self-ownership cannot be combined with a conception of ownership of natural resources
to produce a plausible egalitarian theory, and suggested that this is not obviously so. For,
although joint ownership of natural resources leaves individuals no freedom to do anything
without the consent of others, and equal share ownership does too little to compensate for
unchosen disadvantage, there are plausible accounts of the ownership of natural resources that
avoid the problem of requiring the consent of others to do anything while taking the promotion of
equality more seriously. In particular, accounts that permit unilateral use and appropriation of
natural resources, but which impose tax liabilities on those who appropriate equal to either the
competitive value of the rights claimed or to the full net benefit they realize from appropriation,
seem promising. Although I have not given anything even approximating a full defense of such
accounts, enough has been said, I hope, to show that more investigation into the ownership of
natural resources is needed before egalitarians reject all forms of self-ownership. Needless to say,
without Cohen’s ground-breaking work on the topic, the importance of this investigation for
egalitarianism would not have been adequately recognized.14
References
Ackerman, B. 1980. Social Justice in the Liberal State. New Haven: Yale University Press.
Becker, L. 1977. Property Rights: Philosophical Foundations. Boston: Routledge & Kegan Paul.
Brody, B. 1983. "Redistribution Without Egalitarianism.” Social Philosophy and Policy 1 (1983):
71-87.
Brown, P. 1977. "Food as National Property.” In Food Policy, ed. P. Brown and H. Shue. New
York: The Free Press, 1977.
Christman, J. 1994. The Myth of Property. New York: Oxford University Press.
Cohen, G.A. 1989. "On the Currency of Egalitarian Justice.” Ethics 99 (1989): 906-44.
Colins, H. 1835. Du Pacte Sociale, et de la Liberté Politique considerée comme complément moral
de l’Homme, vol. 2. Paris: Moutardier.
Cunliffe, J. 1996. "Fançois Huet: Intragenerational Liberal; Intergenerational Socialist.”
Unpublished, Department of Law, University of Central England, Birmingham.
---, 1987. "A Mutualist Theory of Exploitation?" In Modern Theories of Exploitation, ed.
Andrew Reeve. London: Sage, 1987.
---, 1988. "The Liberal Rationale of `Rational Socialism'.” Political Studies 36 (1988): 653-662.
---, 1990a. "The Neglected Background of Radical Liberalism; P.E. Dove's Theory of Property.”
History of Political Thought 11 (1990): 467-490.
---, 1990b. "Intergenerational Justice and Productive Resources; A Nineteenth Century Socialist
Debate.” History of European Ideas 12 (1990): 227-238.
Dove, P.E. 1850. The Theory of Human Progression and Natural Probability of a Reign of Justice:
The Science of Politics, Part 1. Edinburgh: Johnstone and Hunten.
21
---, 1854. The Elements of Political Science: The Science of Politics, Part 2. Edinburgh: Johnstone
and Hunten.
Dworkin, R. 1981. "What is Equality? Part 2: Equality of Resources.” Philosophy and Public
Affairs 10 (1981): 283-345.
Fressola, F. 1981. "Liberty and Property." American Philosophical Quarterly 18 (1981): 315-322.
Gauthier, D. 1986. Morals by Agreement. London: Oxford University Press.
George, H. 1879. Progress and Poverty. New York: Robert Schalkenbach Foundation.
---, 1892. A Perplexed Philosopher. New York: Robert Schalkenbach Foundation.
Gibbard, A. 1976. "Natural Property Rights.” Nous 10 (1976): 77-86.
Grunebaum, J. 1987. Private Ownership. New York: Routledge & Kegan Paul.
Haslett, D.W. 1988. "Is Inheritance Justified?" Philosophy and Public Affairs 15 (1988): 122-155.
Huet, F. 1835. Le Regne Social du Christianisme. Paris: Firmin Didot Frères.
Ingram, A. 1994. A Political Theory of Rights. Oxford: Oxford University Press.
Kirzner, I. 1978. “Entrepreneurship, Entitlement, and Economic Justice.” Eastern Economic
Journal 4 (1978). Reprinted in Reading Nozick ed. J. Paul. Oxford: Basil Blackwell.
Kolm, S.C. 1985. Le Contrat Social Libéral (Théorie et Pratique du Libéralisme). Paris: Presses
Universitaires de France.
---, 1986. “L’Allocation des Ressources Naturelles et le Libéralisme.” Revue Economique 37
(1986): 207-241.
---, 1996. Modern Theories of Justice. Cambridge, MA: MIT Press.
Kymlicka, W. 1990. Contemporary Political Philosophy. New York: Oxford.
Levine, A. 1988. “Capitalist Persons.” Social Philosophy and Policy 6 (1988): 39-59.
22
Mack E. 1983. "Distributive Justice and the Tensions of Lockeanism.” Social Philosophy and
Policy 1 (1983): 132-50.
---, 1990. "Self-Ownership and the Right of Property.” Monist 73 (1990): 519-543.
Mavrodes, G. 1974. “Property.” The Personalist 53 (1974): 245-262.
Munzer, S. 1990. A Theory of Property. Cambridge: Cambridge University Press
Narveson, J. 1988. The Libertarian Idea. Philadelphia: Temple University Press.
Nozick, R. 1974. Anarchy, State, and Utopia. New York: Basic Books.
---, 1989. The Examined Life. New York: Simon & Schuster.
Ogilvie, W. 1781. “The Right of Property in Land.” In The Pioneers of Land Reform, ed. M.
Beer. New York: Alfred Knopf, 1920.
Otsuka, M. 1997. “Deontology and Equality: A Lockean Reconciliation.” Unpublished, Department
of Philosophy, University of California at Los Angeles.
Paine, T. 1795. "Agrarian Justice. In The Thomas Paine Reader, ed. M. Foot and I. Kramnick.
Harmondsworth: Penguin Books, 1987.
Rignano, E. 1924. The Social Significance of the Inheritance Tax . New York: Alfred Knopf & Co.
Roemer, J. 1996. Theories of Distributive Justice. Cambridge, MA: Harvard University Press.
Rothbard, M. 1978. For a New Liberty: The Libertarian Manifesto, revised edition. New York:
Libertarian Review Foundation.
---, 1982. The Ethics of Liberty. Atlantic Highlands: Humanities Press.
Sartorius, R. 1984. "Persons and Property.” In Utility and Rights, ed. R.G. Frey. Minneapolis:
University of Minnesota Press, 1984.
Spence, T. 1793. “The Real Rights of Man.” In The Pioneers of Land Reform, ed. M. Beer. New
York: Alfred Knopf, 1920.
23
Spencer, H. 1851. Social Statics. New York: Augustus M. Kelley.
Sreenivasan, G. 1995. The Limits of Lockean Rights in Property. New York: Oxford University
Press.
Steiner, H. 1994. An Essay on Rights. Cambridge, MA: Blackwell Publishers.
Thomson, J.J. 1990. The Realm of Rights. Cambridge, MA: Harvard University Press.
Van Parijs, P. 1995a. Real Freedom for All. New York: Oxford University Press.
---, 1995b. "Free Riding versus Rent Sharing: Why Even David Gauthier Should Support a Basic
Income." In Ethics, Rationality, and Economic Behaviour, ed. F. Farina, F. Hahn, and S.
Vanucci. Oxford: Oxford University Press.
Varian, H. 1975. "Distributive Justice, Welfare Economics, and the Theory of Fairness,"
Philosophy and Public Affairs 4 (1975): 223-47
Waldron, J. 1988. The Right to Private Property. New York: Oxford University Press.
Walras, L. 1936. Etudes d’Economie Sociale, 2nd
edition. Paris: R. Pichon & R. Durand-Auzias.
Wheeler, S. 1980. "Natural Property Rights as Body Rights." Nous 16 (1980): 171-193. Reprinted
in The Main Debate: Communism versus Capitalism, ed. Tibor Machan. New York:
Random House, 1987.
White, S. 1998. “Justice, Community, and the Inequality of Talents.” British Journal of Political
Science, forthcoming.
24
Notes
1 I skip over Cohen's discussion of defenses of self-ownership that appeal to the claims (1) that
people are ends, or (2) that people should not be sacrificed without their consent for the benefit
of others.
2 Cohen assumes, as is standard, that self-ownership, as applied to moral agents, includes
ownership of one’s entire body, but this does not follow from the thesis of self-ownership alone.
It depends on facts about personal identity. If, for example, agents are purely mental beings that
can, but need not, voluntarily occupy and leave bodies, then bodies (in their initial state) are
simply (external) natural resources, just like tracts of land. And even assuming that agents are
necessarily embodied, it does not follow automatically that the agents are identical with their
entire bodies (as opposed to their central nervous system, say). Self-ownership for moral agents
leads to ownership of all of their body parts only if they are identical with their entire bodies. If
correct, this point supplements Cohen’s criticism of the above argument for full self-ownership.
3 In the new Ch. 2 Cohen indirectly (and not explicitly) raises a criticism of the thesis of full self-
ownership that I shall not consider. He there criticizes the principle that whatever arises from a
just situation by just steps is itself just, where the justice of steps (or transactions) is conceived of
procedurally (e.g., fully voluntarily) and not in terms of the ultimate outcomes. He thereby
implicitly questions whether the justice of situations can be captured solely in terms of property
rights. Explaining how this is relevant, and why I think the criticism fails (roughly: because his
concerns are capturable within a property rights framework), would take me too far astray.
25
4 The best known contemporary advocate of the equal division of the natural resource rent is
Steiner. Equal division is also advocated by Van Parijs (1995a), although it is qualified so that
the social fund is divided equally only after compensation has been made for unchosen
disadvantage. Grunebaum (1987) holds the natural resources are jointly owned, but holds that
typically the most sensible decision is to rent out natural resources. If so rented, the rents are
divided equally on his view. Dworkin (1981) develops an equal division approach, but only on
the assumption that there is no differential brute luck. See also the discussion in Kolm (1985,
1986, 1996) and Roemer (1996)
5 The control part of self-ownership is, roughly speaking (ignoring some differences), endorsed
by Christman, A Theory of Property, and (under the name “full act freedom”) Kolm, Modern
Theories of Justice. They both reject full self-ownership.
6 For arguments against the right to make tax-immune gifts or bequests to members of later
generations, see: Ackerman (1980), Haslett (1988), Munzer (1990), Nozick (1989), Rignano
(1924), Sreenivasan (1995), Steiner (1995), Van Parijs (1995a), and Varian (1975).
7 Two fairly plausible conditions on common use are: (1) no one should be made worse off in a
(specified) relevant way compared with non-use (e.g., as when a scarce resource is damaged or
destroyed); and (2) when more people want to use a given resource at a given time than can use
it, each user must pay a fee equal to the competitive value of the use. These conditions are
discussed below with respect to appropriation, but they could also be applied to use.
8 There are many extremely important issues that I am here glossing over. The competitive
values of rights over resources are relative to a set of market conditions, and specifying and
defending a particular set of conditions is crucial. Furthermore, even for a determinate set of
26
initial conditions there can be more than one possible market equilibrium, and hence more than
one possible set of competitive prices for resources.
9 Other early exponents of some form of self-ownership combined with some form of social
ownership of natural resources include: Colins (1835), Dove (1850, 1854), Huet (1835), Ogilvie
(1781), Paine (1795), Spence (1793), Spencer (1851), and Walras (1936). For insightful
discussion of some of these early views, see Cunliffe (1987, 1988, 1990a, 1990b).
10 The idea that the rent from natural resources should be first used to equalize advantage (as
opposed to being equally divided) is endorsed Van Parijs (1995a), Sartorius (1984), and Brown,
(1977).
11 In Ch. 9 (new) Cohen criticizes the claim by Gauthier (1986) that self-ownership does not
include the right to any income in excess of the minimum income necessary to induce one to
supply a service or product. A similar criticism is made by Van Parijs (1995a). As I note in the
text, however, although full ownership of a service or product entitles the owner to the surplus
income, this entitlement may be lost if one uses or appropriates resources that belong in part to
others. Hence, taxation of surplus income is compatible with initial full self-ownership.
12 The following are some loosely related approaches to self-ownership and the ownership of
natural resources: Christman (1994) defends control self-ownership, but denies that agents (even
non-appropriators) have any right to the income they generate. White (1998) accepts control self-
ownership, but argues that agents owe taxes on their income at a rate (varying depending their
talents) that equalizes post-tax potential income. Otsuka (1998) argues that agents are free to
appropriate only as much resources as is compatible with equality of opportunity for welfare (and
thus more talented agents are permitted to appropriate less). Brody (1983) argues that the benefits
27
of appropriation may be taxed to ensure that everyone gets a mutually self-interested fair share of
the benefits. Gibbard (1976) explores full self-ownership in conjunction with joint ownership of
natural resources. This is like the joint ownership model considered by Cohen except that the
non-agreement point is common use rather than non-use. This is a much more plausible model of
joint ownership of natural resources, but it is still implausible because it requires the actual
consent of others for just appropriation (it fails to be unilateralist). He also considers a
hypothetical agreement model.
13 For further discussion of the notion of self-ownership, the ownership of natural resources, and
the ownership of artifacts, see (in addition to the articles already cited): Becker (1977), Fressola
(1981), Ingram (1994), Kymlicka (1990), Mack (1983, 1990), Mavrodes (1974), Narveson
(1988), Thomson (1990), and Waldron (1988).
14 Work on this paper was funded in part by a fellowship grant from the American Council of
Learned Societies. For comments and criticisms I’m indebted to the members of the discussion
groups at Nuffield College, Oxford and at The University of Manchester, Christian Arnsperger,
Jerry Cohen, Tony Ellis, Brad Hooker, Shelly Kagan, Serge Kolm, Michael Lacewing, Trenton
Merricks, Gene Mills, Eric Rakowski, Arthur Ripstein, John Roemer, Hillel Steiner, Andrew
Williams, and two anonymous referees for this journal.